In normal human skin, melanocytes are tightly regulated by surrounding keratinocytes, which determine melanocyte growth, expression of cell surface molecules, and cell shape. This dominance of keratinocytes over the melanocytes is lost during transformation, allowing melanoma cells to invade into the dermis. By studying cell-cell communication between normal melanocytes and keratinocytes, we are beginning to understand how melanoma cells have escaped from keratinocytes' control. We now expand for normal human skin the homeostatic concept of growth, differentiation, and survival or death and include stem cells as a continuing source for differentiated, pigment-producing melanocytes. As model systems we use melanocytes differentiated from human embryonic stem cells (NIH cell line codes WA01 and WA09) and progenitor cells of the human hair follicle. The working hypothesis in the first aim is that cadherin-mediated cell-cell interactions determine self-renewal of multi-potent progenitor, oligo-potent precursor and committed melanocyte stem cells whereas cell-matrix interactions drive differentiation. Using a three-dimensional model of normal human skin with a 'dermis' of fibroblasts embedded in collagen and an 'epidermis' with melanocytes and multi-layered keratinocytes, we will determine whether proper positioning of mature melanocytes at the basement membrane is controlled by integrin D6B1 and matricellular protein CCN3 and whether improper activation of cell-cell signaling genes such as Notchl leads to transformation. Our hypothesis is that melanocyte precursor and stem cells are more prone to transformation than mature, pigment-producing melanocytes. Using culture conditions developed for human embryonic stem cells, we have for aim 2 isolated melanoma cells from patients' lesions that have stem cell-like characteristics and differentiate into melanogenic, adipogenic, chondrogenic, and osteogenic cell lineages. These 'melanoma stem cells' represent a minor tumorigenic population and are characterized by the expression of the B cell marker CD20. We will determine whether and how cell-cell and cell-matrix interactions regulate the CD20- defined melanoma stem cell population and how CD20 can be optimally targeted for melanoma therapy. These studies will demonstrate how dysregulation of homeostasis results in tumor formation and targeting subpopulations of malignant cells leads to successful therapy.